CN111858458B - Method, device, system, equipment and medium for adjusting interconnection channel - Google Patents
Method, device, system, equipment and medium for adjusting interconnection channel Download PDFInfo
- Publication number
- CN111858458B CN111858458B CN202010568700.6A CN202010568700A CN111858458B CN 111858458 B CN111858458 B CN 111858458B CN 202010568700 A CN202010568700 A CN 202010568700A CN 111858458 B CN111858458 B CN 111858458B
- Authority
- CN
- China
- Prior art keywords
- node
- nodes
- competition
- value
- performance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000010801 machine learning Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 5
- 239000000126 substance Substances 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 20
- 238000013468 resource allocation Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 8
- 241000170489 Upis Species 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/163—Interprocessor communication
- G06F15/17—Interprocessor communication using an input/output type connection, e.g. channel, I/O port
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
Abstract
The embodiment of the invention discloses a method, a device, a system, equipment and a medium for adjusting an interconnection channel, which are used for calculating a competition value of each node according to the resource utilization rate and the type parameter of each node; the competition value reflects the degree of the requirement of the node on the resource, and for the node with a larger competition value, if only processing by a single node is relied on, the processing performance of the single node is limited, while the resources of other nodes with smaller competition values are not fully utilized. Therefore, the callable nodes with the performance parameters meeting the preset conditions can be selected; and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node. By adjusting the interconnection channels among the nodes according to the competition values and the performance parameters of the nodes, the tasks of the nodes with higher resource requirements can be transferred to the nodes with lower resource requirements, the resources of the nodes can be fully called, the purpose of balanced resource allocation is achieved, and the resource utilization rate is improved.
Description
Technical Field
The present invention relates to the field of cloud processing technologies, and in particular, to a method, an apparatus, a system, a device, and a computer-readable storage medium for adjusting an interconnection channel.
Background
Cloud has become a hot word in the emerging information technology era, and cloud computing and cloud processing become important guarantees of various enterprises for market competition and technical competition. The recent epidemic situation further promotes the emergence of new working modes such as cloud office and the like, and also puts higher requirements on data processing, network maintenance and maximum utilization of server resources. In the face of the pressure and the challenge of the server to process larger information storm, the reasonable utilization of server resources and the reasonable allocation of resource competition become a key point of concern during the development of the server.
A single Central Processing Unit (CPU) has limited Processing capability, and a cloud platform generally uses a plurality of CPUs to provide services. A super Path Interconnect (UPI) may implement interconnection between multiple CPUs.
Fig. 1 shows a UPI high-speed interconnection topology of a 2-way server platform in the prior art, and two CPUs are interconnected through 4 UPIs. However, the interconnection mode is complex, and in the face of simple calculation, many unreasonable situations exist in resource scheduling. In the face of complex high-density computation, resource allocation and reasonable resource competition of a multi-UPI topology have challenges, and resource allocation is unreasonable.
Therefore, how to achieve the balanced allocation of resources and improve the resource utilization rate is a problem to be solved by the technical personnel in the field.
Disclosure of Invention
Embodiments of the present invention provide a method, an apparatus, a system, a device, and a computer-readable storage medium for adjusting an interconnection channel, which can implement balanced allocation of resources and improve resource utilization.
To solve the foregoing technical problem, an embodiment of the present invention provides a method for adjusting an interconnection channel, including:
calculating the competition value of each node according to the resource utilization rate and the type parameter of each node;
selecting out the callable nodes with the performance parameters meeting the preset conditions;
and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node.
Optionally, the selecting the callable node whose performance parameter meets the preset condition includes:
acquiring performance parameters of each node; the performance parameters comprise CPU utilization rate, memory occupancy rate and I/O performance indexes;
judging whether the performance parameters of each node exceed a preset threshold value or not; each type of performance parameter has a corresponding preset threshold value;
And taking the nodes of which the performance parameters do not exceed the preset threshold value as the callable nodes.
Optionally, after the obtaining the performance parameters of each node, the method further includes:
when a target node with the performance parameter exceeding a preset threshold exists, a pause identifier is set for the target node to stop distributing tasks to the target node.
Optionally, the calculating, according to the resource utilization rate and the type parameter of each node, a contention value of each node includes:
calculating the competition value I of the current node according to the following formulaCPU，
Wherein, c0And c1Representing type parameters, X, derived based on machine learning training for characterizing the performance of the current nodeiAnd the resource utilization rate of the ith task of the current node obtained based on machine learning training is shown, and n is the total number of tasks of the current node.
Optionally, the creating, according to the contention value and the performance parameter of each node, a corresponding number of interconnection channels between the current node and the callable node includes:
selecting nodes to be migrated with the competition value exceeding a preset upper limit value from all the nodes; selecting a node to be received with a competition value smaller than a preset lower limit value from all the callable nodes;
determining the number of channels to be opened between the node to be migrated and the node to be received according to the performance parameters of the node to be migrated and the performance parameters of the node to be received;
And establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
Optionally, after an interconnection channel is created between the node to be migrated and the node to be received, the method further includes:
and when detecting that the competition value of the node to be migrated is smaller than a preset lower limit value, closing an interconnection channel between the node to be migrated and the node to be received.
The embodiment of the invention also provides an adjusting device of the interconnection channel, which comprises a computing unit, a selecting unit and a creating unit;
the computing unit is used for computing the competition value of each node according to the resource utilization rate and the type parameter of each node;
the selection unit is used for selecting the callable nodes with the performance parameters meeting the preset conditions;
and the creating unit is used for creating the corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes.
Optionally, the selecting unit includes an acquiring subunit, a judging subunit and a serving subunit;
the acquiring subunit is configured to acquire a performance parameter of each node; the performance parameters comprise CPU utilization rate, memory occupancy rate and I/O performance indexes;
The judging subunit is configured to judge whether the performance parameter of each node exceeds a preset threshold; each type of performance parameter has a corresponding preset threshold value;
and the serving as a subunit, configured to serve, as an invocable node, a node whose performance parameter does not exceed a preset threshold.
Optionally, a setting unit is further included;
the setting unit is used for setting a pause identifier for a target node when the target node with the performance parameter exceeding a preset threshold exists so as to stop distributing tasks to the target node.
Optionally, the calculating unit is specifically configured to calculate a contention value I of the current node according to the following formulaCPU，
Wherein, c0And c1Representing type parameters, X, derived based on machine learning training for characterizing the performance of the current nodeiRepresenting work derived from machine learning trainingThe resource utilization rate of the ith task of the previous node, and n represents the total number of tasks of the current node.
Optionally, the creating unit includes a selecting subunit and a creating subunit;
the selecting subunit is configured to select, from all the nodes, a node to be migrated whose contention value exceeds a preset upper limit value; selecting a node to be received with a competition value smaller than a preset lower limit value from all the callable nodes;
The creating subunit is configured to determine, according to the performance parameter of the node to be migrated and the performance parameter of the node to be received, the number of channels that need to be opened between the node to be migrated and the node to be received; and establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
Optionally, a closing unit is further included;
and the closing unit is configured to close an interconnection channel between the node to be migrated and the node to be received when it is detected that the contention value of the node to be migrated is smaller than a preset lower limit value.
The embodiment of the invention also provides an adjusting system of the interconnection channel, which comprises a plurality of nodes and a channel controller; according to a structure that a plurality of nodes are connected in series to form a circle, a fixed channel is arranged between every two adjacent nodes; the channel controller is in specific communication connection with the plurality of nodes respectively;
the channel controller is used for calculating the competition value of each node according to the resource utilization rate and the type parameter of each node; selecting out the callable nodes with the performance parameters meeting the preset conditions; and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node.
An embodiment of the present invention further provides an apparatus for adjusting an interconnection channel, including:
a memory for storing a computer program;
a processor configured to execute the computer program to implement the steps of the method for adjusting an interconnection channel according to any one of the above-mentioned items.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the step of the method for adjusting an interconnection channel is implemented as any one of the above.
According to the technical scheme, the competition value of each node is calculated according to the resource utilization rate and the type parameter of each node; the competition value reflects the degree of the resource demand of the node, and the larger the competition value of the node is, the more resources are required by the node to execute the current task. For the nodes with larger competition values, if only processing of a single node is relied on, the processing performance of the single node is limited, and the resources of other nodes with smaller competition values are not fully utilized. Therefore, in the technical scheme, the callable nodes with the performance parameters meeting the preset conditions can be selected; and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes. By adjusting the interconnection channel between the nodes according to the competition value and the performance parameter of each node, the task of the node with higher resource demand can be transferred to the node with lower resource demand, so that the resource of each node can be fully called, the aim of balanced resource distribution is fulfilled, and the resource utilization rate is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention, the drawings required for the embodiments will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained by those skilled in the art without inventive effort.
Fig. 1 is a UPI high-speed interconnection topology diagram of a 2-way server platform provided in the prior art;
fig. 2 is a flowchart of an interconnection channel adjusting method according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an adjusting device of an interconnection channel according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an interconnection channel adjustment system according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an adjusting apparatus for an interconnection channel according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without any creative work belong to the protection scope of the present invention.
In order that those skilled in the art will better understand the disclosure, reference will now be made in detail to the embodiments of the disclosure as illustrated in the accompanying drawings.
Next, a method for adjusting an interconnection channel according to an embodiment of the present invention is described in detail. Fig. 2 is a flowchart of a method for adjusting an interconnection channel according to an embodiment of the present invention, where the method includes:
s201: and calculating the competition value of each node according to the resource utilization rate and the type parameter of each node.
In the embodiment of the invention, the establishment and the closing of the channel among the multiple nodes can be realized through the channel controller. In order to satisfy the basic communication function between the nodes, a plurality of nodes may be connected in series to form a ring, and a fixed channel may be provided between each adjacent two nodes. The channel controller is in specific communication connection with the nodes respectively, and a new interconnection channel can be established between the nodes according to resource requirements.
The way of calculating the contention value by each node is the same, and in the embodiment of the present invention, one node is taken as an example for description.
There may be multiple tasks that need to be processed on each node, each task having its corresponding resource utilization. In the embodiment of the invention, the resource utilization rate of the ith task of the current node can be acquired in a machine learning mode.
Considering that the configuration of the core hardware of the server such as CPU, memory, I/O and the like of different nodes is different, each node has a corresponding type parameter. In practical application, the type parameters for characterizing the performance of the current node can be obtained based on machine learning training.
The type parameter of the node and the resource utilization rate of each task on the node are important factors influencing the performance of the node, and in the concrete implementation, the competition value I of the current node can be calculated according to the following formulaCPU，
Wherein, c0And c1Representing type parameters, X, derived based on machine learning training for characterizing the performance of the current nodeiAnd the resource utilization rate of the ith task of the current node obtained based on machine learning training is shown, and n is the total number of tasks of the current node.
S202: and selecting the callable nodes with the performance parameters meeting the preset conditions.
The performance parameters of the node refer to parameters affecting the processing capability of the node, and may include CPU utilization, memory occupancy, I/O performance index, and the like. The I/O performance indicators may include I/O usage and I/O saturation, among others.
The preset condition may be a condition that is satisfied when the node is able to receive a new task. In specific implementation, the performance parameters such as the CPU utilization rate, the memory occupancy rate, the I/O performance index, and the like may be comprehensively evaluated, or different types of performance parameters such as the CPU utilization rate, the memory occupancy rate, the I/O performance index, and the like may be individually evaluated.
Taking individual evaluation of different types of performance parameters as an example, in specific implementation, it may be determined whether the performance parameters of each node exceed a preset threshold, and nodes whose performance parameters do not exceed the preset threshold are taken as callable nodes. When a target node with the performance parameter exceeding the preset threshold exists, it indicates that the target node cannot process a new task, and at this time, a pause identifier may be set for the target node, so that the server system stops allocating tasks to the target node.
In practical applications, the critical hardware I/O performance index may be monitored by an IC such as a Complex Programmable Logic Device (CPLD), a Baseboard Management Controller (BMC), or a PCA 9555. The memory occupancy rate is monitored through a Basic Input Output System (BIOS) or software dedicated to a memory manufacturer.
Each type of performance parameter has its own corresponding preset threshold. The value of the preset threshold can be set according to the actual demand of the user and the accumulation of the actual measurement result.
S203: and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes.
The competition value reflects the degree of the resource demand of the node, and the larger the competition value of the node is, the more resources are required by the node to execute the current task.
The current node may be any one of all nodes. In the embodiment of the invention, the task on the node with a larger competition value can be transferred to the node with a smaller competition value by establishing the interconnection channel between the nodes, so that the balanced use of each node resource is realized, and the utilization rate of the node resource is improved.
In a specific implementation, a node to be migrated whose contention value exceeds a preset upper limit value may be selected from all nodes. When the contention value of the node exceeds the preset upper limit value, it indicates that the node has more tasks to be processed currently, and needs to occupy more resources, and if only processing by a single node is relied on, the processing performance of the single node is limited, and the resources of other nodes with smaller contention values are not fully utilized. Therefore, in the embodiment of the present invention, a node whose contention value is smaller than the preset lower limit value may be selected as a node to be received.
The values of the preset upper limit value and the preset lower limit value can be set according to actual requirements, and are not limited herein.
In practical application, the number of channels to be opened between the node to be migrated and the node to be received can be determined according to the performance parameters of the node to be migrated and the performance parameters of the node to be received; and establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
The performance parameters of the node reflect the current processing capability of the node. The performance parameters include CPU utilization, memory occupancy, I/O performance index, and the like. Taking a node to be received as an example, when the CPU utilization rate, the memory occupancy rate, or the I/O utilization rate and the I/O saturation of the node to be received are low, it indicates that the resources of the node to be received are not fully used, and the node to be received has a high processing capability, and at this time, a plurality of interconnection channels may be created between the node to be migrated and the node to be received.
It should be noted that, limited by the performance requirements of the CPUs in the nodes, each CPU can support parallel operation of at most 6 interconnection channels, and therefore, when creating interconnection channels between nodes, the number of channels currently opened by the nodes should be fully considered.
According to the technical scheme, the competition value of each node is calculated according to the resource utilization rate and the type parameter of each node; the competition value reflects the degree of the requirement of the node on the resources, and the larger the competition value of the node is, the more resources are required for the node to execute the current task. For the nodes with larger competition values, if only processing of a single node is relied on, the processing performance of the single node is limited, and the resources of other nodes with smaller competition values are not fully utilized. Therefore, in the technical scheme, the callable nodes with the performance parameters meeting the preset conditions can be selected; and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes. By adjusting the interconnection channel between the nodes according to the competition value and the performance parameter of each node, the task of the node with higher resource demand can be transferred to the node with lower resource demand, so that the resource of each node can be fully called, the aim of balanced resource distribution is fulfilled, and the resource utilization rate is improved.
In the embodiment of the invention, the interconnection channel can be established between the node with a larger competition value and the node with a smaller competition value to realize the migration of the node task, thereby improving the utilization rate of the node resource. However, as time goes on, the amount of tasks required to be executed on each node changes, so that the interconnection channel created between the nodes may not be suitable for the current task requirements of the nodes. Therefore, in the embodiment of the present invention, the contention value of each node may be detected in real time or periodically, and when it is detected that the contention value of the node to be migrated is smaller than the preset lower limit value, the interconnection channel between the node to be migrated and the node to be received is closed.
The interconnection channels between the nodes are dynamically adjusted according to the competition values of the nodes, so that the number of the interconnection channels between the nodes is more suitable for the actual requirements of the nodes, and the balanced distribution of the resources of the whole system is realized.
Fig. 3 is a schematic structural diagram of an adjusting apparatus of an interconnection channel according to an embodiment of the present invention, including a calculating unit 31, a selecting unit 32, and a creating unit 33;
a calculating unit 31, configured to calculate a contention value of each node according to the resource utilization rate and the type parameter of each node;
A selecting unit 32, configured to select a callable node whose performance parameter meets a preset condition;
the creating unit 33 is configured to create a corresponding number of interconnection channels between the current node and the callable node according to the contention value and the performance parameter of each node.
Optionally, the selecting unit comprises an acquiring subunit, a judging subunit and a serving subunit;
an obtaining subunit, configured to obtain performance parameters of each node; the performance parameters comprise CPU utilization rate, memory occupancy rate and I/O performance indexes;
the judging subunit is used for judging whether the performance parameters of each node exceed a preset threshold value or not; each type of performance parameter has a corresponding preset threshold value;
and the sub-unit is used for taking the nodes of which the performance parameters do not exceed the preset threshold value as the callable nodes.
Optionally, the system further comprises a setting unit;
and the setting unit is used for setting a pause identifier for the target node to stop distributing the task to the target node when the target node with the performance parameter exceeding the preset threshold exists.
Optionally, the calculating unit is specifically configured to calculate the contention value I of the current node according to the following formulaCPU，
Wherein, c0And c1Representing type parameters, X, derived based on machine learning training for characterizing the performance of the current node iAnd the resource utilization rate of the ith task of the current node obtained based on machine learning training is shown, and n is the total number of tasks of the current node.
Optionally, the creating unit includes a selecting subunit and a creating subunit;
the selecting subunit is used for selecting the nodes to be migrated with the competition values exceeding a preset upper limit value from all the nodes; selecting a node to be received with a competition value smaller than a preset lower limit value from all the nodes capable of being called;
the creating subunit is used for determining the number of channels to be opened between the node to be migrated and the node to be received according to the performance parameters of the node to be migrated and the performance parameters of the node to be received; and establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
Optionally, a closing unit is further included;
and the closing unit is used for closing the interconnection channel between the node to be migrated and the node to be received when detecting that the competition value of the node to be migrated is smaller than the preset lower limit value.
The description of the features in the embodiment corresponding to fig. 3 may refer to the related description of the embodiment corresponding to fig. 2, and is not repeated here.
According to the technical scheme, the competition value of each node is calculated according to the resource utilization rate and the type parameter of each node; the competition value reflects the degree of the requirement of the node on the resources, and the larger the competition value of the node is, the more resources are required for the node to execute the current task. For the nodes with larger competition values, if only processing of a single node is relied on, the processing performance of the single node is limited, and the resources of other nodes with smaller competition values are not fully utilized. Therefore, in the technical scheme, the callable nodes with the performance parameters meeting the preset conditions can be selected; and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes. By adjusting the interconnection channel between the nodes according to the competition value and the performance parameter of each node, the task of the node with higher resource demand can be transferred to the node with lower resource demand, so that the resource of each node can be fully called, the aim of balanced resource distribution is fulfilled, and the resource utilization rate is improved.
Fig. 4 is a schematic structural diagram of an interconnection channel adjustment system according to an embodiment of the present invention, including multiple nodes 41 and a channel controller 42; according to the structure that a plurality of nodes 41 are connected in series to form a circle, a fixed channel is arranged between every two adjacent nodes 41; the channel controller 42 is in communication connection with the plurality of nodes 41;
a channel controller 42, configured to calculate a contention value of each node 41 according to the resource utilization rate and the type parameter of each node 41; selecting the callable nodes 41 with the performance parameters meeting the preset conditions; and dynamically adjusting the interconnection channel between the current node 41 and each callable node 41 according to the competition value of each node 41.
The interconnection channel between nodes refers to the connection between CPUs in the nodes, and fig. 4 is a schematic diagram of a connection relationship by taking 4 CPUs as an example, and one CPU belongs to one node. In fig. 4, the CPU0, the CPU1, the CPU2, and the CPU3 are provided with their own channels in a serial connection manner, and taking the CPU0 as an example, the CPU0 establishes 2-way interconnection channels (UPIs) with the CPU1 and the CPU3, respectively. P0-P5 in each CPU refer to the ports of the CPU. The dashed lines between the CPUs in FIG. 4 refer to interconnect channels that may be dynamically created or closed.
The description of the features in the embodiment corresponding to fig. 4 can refer to the related description of the embodiment corresponding to fig. 2, and is not repeated here.
According to the technical scheme, the competition value of each node is calculated according to the resource utilization rate and the type parameter of each node; the competition value reflects the degree of the resource demand of the node, and the larger the competition value of the node is, the more resources are required by the node to execute the current task. For the nodes with larger competition values, if only processing of a single node is relied on, the processing performance of the single node is limited, and the resources of other nodes with smaller competition values are not fully utilized. Therefore, in the technical scheme, the callable nodes with the performance parameters meeting the preset conditions can be selected; and establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node. By adjusting the interconnection channel between the nodes according to the competition value and the performance parameter of each node, the task of the node with higher resource demand can be transferred to the node with lower resource demand, so that the resource of each node can be fully called, the aim of balanced resource distribution is fulfilled, and the resource utilization rate is improved.
Fig. 5 is a schematic structural diagram of an adjusting apparatus 50 for an interconnection channel according to an embodiment of the present invention, including:
A memory 51 for storing a computer program;
a processor 52 for executing a computer program to implement the steps of the method for adjusting an interconnect channel according to any of the embodiments described above.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method for adjusting an interconnection channel according to any of the above embodiments are implemented.
The method, the apparatus, the system, the device and the computer readable storage medium for adjusting an interconnection channel according to the embodiments of the present invention are described in detail above. The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Those of skill would further appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the components and steps of the various examples have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Claims (8)
1. A method for adjusting an interconnection channel, comprising:
Calculating the competition value of each node according to the resource utilization rate and the type parameter of each node;
selecting out the callable nodes with the performance parameters meeting preset conditions;
establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node;
the calculating the competition value of each node according to the resource utilization rate and the type parameter of each node comprises:
calculating the competition value of the current node according to the following formula
，
Wherein, the first and the second end of the pipe are connected with each other,
and
representing type parameters for characterizing the performance of the current node based on machine learning training,
representing a current node derived based on machine learning training
The resource utilization rate of the individual tasks,
representing the total number of tasks of the current node;
the creating a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes comprises:
selecting nodes to be migrated with the competition value exceeding a preset upper limit value from all the nodes; selecting a node to be received with a competition value smaller than a preset lower limit value from all the callable nodes;
determining the number of channels to be opened between the node to be migrated and the node to be received according to the performance parameters of the node to be migrated and the performance parameters of the node to be received;
And establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
2. The method of claim 1, wherein the selecting the callable node with the performance parameter meeting the predetermined condition comprises:
acquiring performance parameters of each node; the performance parameters comprise CPU utilization rate, memory occupancy rate and I/O performance indexes;
judging whether the performance parameters of each node exceed a preset threshold value or not; each type of performance parameter has a corresponding preset threshold value;
and taking the nodes of which the performance parameters do not exceed the preset threshold value as the callable nodes.
3. The method of claim 2, further comprising, after the obtaining performance parameters of each node:
and when a target node with the performance parameter exceeding a preset threshold value exists, setting a pause identifier for the target node to stop distributing tasks to the target node.
4. The method according to claim 1, further comprising, after creating a interconnect channel between the node to be migrated and the node to be received:
and when detecting that the competition value of the node to be migrated is smaller than a preset lower limit value, closing an interconnection channel between the node to be migrated and the node to be received.
5. The adjusting device of the interconnection channel is characterized by comprising a computing unit, a selecting unit and a creating unit;
the computing unit is used for computing the competition value of each node according to the resource utilization rate and the type parameter of each node;
the selection unit is used for selecting the callable nodes with the performance parameters meeting the preset conditions;
the creation unit is used for creating a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes;
the calculating the competition value of each node according to the resource utilization rate and the type parameter of each node comprises:
calculating the competition value of the current node according to the following formula
，
Wherein, the first and the second end of the pipe are connected with each other,
and
representing type parameters for characterizing the performance of the current node based on machine learning training,
representing a current node derived based on machine learning training
The resource utilization rate of the individual tasks,
representing the total number of tasks of the current node;
the creating a corresponding number of interconnection channels between the current node and the callable node according to the competition values and the performance parameters of the nodes comprises:
selecting nodes to be migrated with the competition value exceeding a preset upper limit value from all the nodes; selecting a node to be received with a competition value smaller than a preset lower limit value from all the callable nodes;
Determining the number of channels required to be opened between the node to be migrated and the node to be received according to the performance parameters of the node to be migrated and the performance parameters of the node to be received;
and establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
6. The adjusting system of the interconnection channel is characterized by comprising a plurality of nodes and a channel controller; according to a structure that a plurality of nodes are connected in series to form a circle, a fixed channel is arranged between every two adjacent nodes; the channel controller is in specific communication connection with the plurality of nodes respectively;
the channel controller is used for calculating the competition value of each node according to the resource utilization rate and the type parameter of each node; selecting out the callable nodes with the performance parameters meeting the preset conditions; establishing a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node;
the calculating the competition value of each node according to the resource utilization rate and the type parameter of each node comprises:
calculating the competition value of the current node according to the following formula
，
Wherein the content of the first and second substances,
and
representing type parameters for characterizing the performance of the current node based on machine learning training, 
Representing the current node based on machine learning training
The resource utilization rate of the individual tasks is,
indicating the total number of tasks of the current node;
the creating of a corresponding number of interconnection channels between the current node and the callable node according to the competition value and the performance parameter of each node comprises:
selecting nodes to be migrated with the competition value exceeding a preset upper limit value from all the nodes; selecting a node to be received with a competition value smaller than a preset lower limit value from all the callable nodes;
determining the number of channels to be opened between the node to be migrated and the node to be received according to the performance parameters of the node to be migrated and the performance parameters of the node to be received;
and establishing an interconnection channel between the node to be migrated and the node to be received according to the number of the channels.
7. An interconnect channel tuning apparatus, comprising:
a memory for storing a computer program;
processor for executing said computer program for implementing the steps of the method of tuning an interconnect channel according to any of claims 1 to 4.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for adjusting an interconnect channel according to any one of claims 1 to 4.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010568700.6A CN111858458B (en) | 2020-06-19 | 2020-06-19 | Method, device, system, equipment and medium for adjusting interconnection channel |
| PCT/CN2021/071202 WO2021253817A1 (en) | 2020-06-19 | 2021-01-12 | Interconnection channel adjusting method, apparatus, system and device, and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010568700.6A CN111858458B (en) | 2020-06-19 | 2020-06-19 | Method, device, system, equipment and medium for adjusting interconnection channel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111858458A CN111858458A (en) | 2020-10-30 |
| CN111858458B true CN111858458B (en) | 2022-05-24 |
Family
ID=72987746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010568700.6A Active CN111858458B (en) | 2020-06-19 | 2020-06-19 | Method, device, system, equipment and medium for adjusting interconnection channel |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111858458B (en) |
| WO (1) | WO2021253817A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111858458B (en) * | 2020-06-19 | 2022-05-24 | 苏州浪潮智能科技有限公司 | Method, device, system, equipment and medium for adjusting interconnection channel |
| CN117519951A (en) * | 2024-01-04 | 2024-02-06 | 深圳博瑞天下科技有限公司 | Real-time data processing method and system based on message center |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110255868A1 (en) * | 2009-07-07 | 2011-10-20 | Zte Corporation | method and a device of service migration in an automatically switched optical network |
| CN102369511A (en) * | 2011-09-01 | 2012-03-07 | 华为技术有限公司 | Resource removing method, device and system |
| CN107220125A (en) * | 2017-05-27 | 2017-09-29 | 郑州云海信息技术有限公司 | A kind of cloud resource dispatching method and device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101916239B (en) * | 2010-08-27 | 2011-09-28 | 上海交通大学 | Method for enhancing communication speed of on-chip multiprocessor |
| CN105335229B (en) * | 2014-07-25 | 2020-07-07 | 新华三技术有限公司 | Scheduling method and device of service resources |
| CN105528199B (en) * | 2014-09-30 | 2019-04-19 | 华为技术有限公司 | A kind of processing method and processing device of node |
| CN106255228B (en) * | 2016-04-27 | 2020-09-11 | 北京智谷睿拓技术服务有限公司 | Method for establishing network connection, terminal equipment and node equipment |
| US10503684B2 (en) * | 2016-07-01 | 2019-12-10 | Intel Corporation | Multiple uplink port devices |
| US10728311B2 (en) * | 2017-02-16 | 2020-07-28 | Intel Corporation | Device, system and method for adaptive payload compression in a network fabric |
| US20200042608A1 (en) * | 2018-08-01 | 2020-02-06 | EMC IP Holding Company LLC | Distributed file system load balancing based on available node capacity |
| CN111858458B (en) * | 2020-06-19 | 2022-05-24 | 苏州浪潮智能科技有限公司 | Method, device, system, equipment and medium for adjusting interconnection channel |
-
2020
- 2020-06-19 CN CN202010568700.6A patent/CN111858458B/en active Active
-
2021
- 2021-01-12 WO PCT/CN2021/071202 patent/WO2021253817A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110255868A1 (en) * | 2009-07-07 | 2011-10-20 | Zte Corporation | method and a device of service migration in an automatically switched optical network |
| CN102369511A (en) * | 2011-09-01 | 2012-03-07 | 华为技术有限公司 | Resource removing method, device and system |
| CN107220125A (en) * | 2017-05-27 | 2017-09-29 | 郑州云海信息技术有限公司 | A kind of cloud resource dispatching method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111858458A (en) | 2020-10-30 |
| WO2021253817A1 (en) | 2021-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112153700B (en) | Network slice resource management method and equipment | |
| CN104102543B (en) | The method and apparatus of adjustment of load in a kind of cloud computing environment | |
| US8463971B2 (en) | Approach for distributing interrupts from high-interrupt load devices | |
| JP5664098B2 (en) | Composite event distribution apparatus, composite event distribution method, and composite event distribution program | |
| US10909018B2 (en) | System and method for end-to-end application root cause recommendation | |
| US7467291B1 (en) | System and method for calibrating headroom margin | |
| US20070043970A1 (en) | Approach for managing interrupt load distribution | |
| US20070250630A1 (en) | Method and a system of generating and evaluating potential resource allocations for an application | |
| CN107426003B (en) | Fault detection method and device | |
| CN111858458B (en) | Method, device, system, equipment and medium for adjusting interconnection channel | |
| WO2015101091A1 (en) | Distributed resource scheduling method and device | |
| CN112506643A (en) | Load balancing method and device of distributed system and electronic equipment | |
| CN109587072A (en) | Distributed system overall situation speed limiting system and method | |
| US11669374B2 (en) | Using machine-learning methods to facilitate experimental evaluation of modifications to a computational environment within a distributed system | |
| Nasim et al. | Optimizing virtual machine consolidation in virtualized datacenters using resource sensitivity | |
| CN111418187A (en) | Scalable statistics and analysis mechanism in cloud networks | |
| CN114327824A (en) | Method and device for determining service host and electronic equipment | |
| CN111800291B (en) | Service function chain deployment method and device | |
| CN109992408B (en) | Resource allocation method, device, electronic equipment and storage medium | |
| CN114500578A (en) | Load balancing scheduling method and device for distributed storage system and storage medium | |
| US11374869B2 (en) | Managing bandwidth based on user behavior | |
| Wei et al. | Towards multi-resource physical machine provisioning for IaaS clouds | |
| Marinescu et al. | Distributed hierarchical control versus an economic model for cloud resource management | |
| CN111598390A (en) | Server high availability evaluation method, device, equipment and readable storage medium | |
| JP7235296B2 (en) | SESSION MANAGEMENT METHOD, SESSION MANAGEMENT DEVICE, AND PROGRAM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |